Relative adrenal insufficiency is a risk factor and an endotype of sepsis - A proof of concept study to support a precision medicine approach for glucocorticoid sepsis therapy

Rational25-60% of septic patients experience relative adrenal insufficiency (RAI) and glucocorticoid (GC) is frequently used in septic patients. However, the efficacy of GC therapy and whether the GC therapy should be based on the status of RAI are highly controversial. Critical barriers include technical limitations in properly identifying RAI in septic patients and a lack of RAI animal model. ObjectivesWe established a new RAI animal model to test our hypothesis that precision medicine approach should be used for GC sepsis therapy - only applying GC to a subgroup of septic mice with RAI. MethodsWe generated SF1CreSR-BIfl/fl conditional knockout mice. The mice exhibited specific depletion of SR-BI expression in adrenal gland, resulting in a lack of production of inducible GC in response to ACTH stimulation or sepsis, but the mice had normal basal GC levels. Mice were treated with cecal ligation and puncture to develop sepsis. Mice were also supplemented with or without GC to study the effect of GC in sepsis therapy. Plasma and organs were collected for biochemical assays. BODIPY FL-conjugated Escherichia coli was used for phagocytosis assay. Macrophages were used to study effects of GC on inflammatory responses. Measurements and Main ResultsUsing SF1CreSR-BIfl/fl mice as a RAI model, we found that mice with RAI were susceptible to CLP-induced sepsis compared to controls (6.7% survival in SF1CreSR-BIfl/fl mice versus 86.4% in SR-BIfl/fl mice; p=0.0001). Supplementation of hydrocortisone significantly improved survival in CLP-treated SF1CreSR-BIfl/fl mice. Surprisingly, wild type mice receiving GC treatment exhibited significantly less survival compared to wild type mice without GC treatment. We further found that, in contrast to wild type mice which displayed a well-controlled systemic inflammatory response, the mice with RAI featured a persisted systemic response as shown by high levels of plasma inflammatory cytokines/chemokines 20 hours post CLP, and supplementation of GC kept the inflammatory response under control. In vitro analysis revealed that stress level of GC is required to suppress inflammatory response through modulating MAPK signaling in macrophages. ConclusionsWe demonstrate that RAI is a risk factor and an endotype for sepsis, and GC treatment benefits mice with RAI but harms mice without RAI. We further demonstrate that inducible GC functions to keep the systemic inflammatory response under control through modulating MAPK signaling, but mice with RAI lose such protection and supplementation of GC regains the protection. Our study provides a proof of concept to support the use of a precision medicine approach for sepsis therapy - selectively applying GC therapy for a subgroup of patients with RAI.